To fully understand the pathogenic process of human viral diseases animal models are needed. We have developed the SCID-hu mouse as an in vivo system which demonstrates HIV-l-induced pathology closely resembling that seen in human disease. The SCID-hu system employs immunodeficient mice as hosts for human fetal hematopoietic tissue (thymus/liver) grafts, which can be maintained for up to a year. We recently found that these human Thy/Liv grafts in HIV-l-infected SCID-hu mice undergo marked depletion of CD4-positive cells in a manner characteristic of that seen in human HIV-l infection. Thus, we have a model uniquely suited to the study of HIV-I pathogenesis in vivo which avoids the artifacts common to in vitro systems. We will use this model to investigate the molecular mechanisms used by HIV-l to deplete human thymocytes in vivo and will also introduce viruses mutated in the auxiliary genes, nef, vif, vpu, and vpr, to determine the role these genes play in the pathogenic process. Our preliminary data indicate that nef mutants of HIV-l display attenuated growth and cytopathic properties when compared to wild-type virus in this system. Thus, the SCID-hu mouse appears to function as an in vivo system to assess attenuation of HIV-l strains. Site-directed mutations in previously identified "active" regions and/or deletion mutants of the nef gene will be introduced into infectious virus and used in the SCID-hu system to determine the critical regions of nef in vivo. These experiments should increase our knowledge of how HIV-l depletes cells in a lymphoid organ, and provide valuable information regarding the role of viral auxiliary genes on virus replication in vivo. These studies will also help to develop the SCID-hu system as a means of screening "attenuated" viruses which may augment the development of a live attenuated vaccine against AIDS.